In order to design trustworthy hydraulic prevention structures, calculating proper flood discharge by using runoff analysis is crucial. To do so, Clark unit hydrograph (UH), which is nicely applicable in ungaged basin, is widely used these days. The parameters of Clark UH have to be obtained with enough actual measurement, but this is unrealistic in real world. Therefore, those parameters are usually obtained by optimization of concentration time (TC) and storage constant (K) using empirical formula.
Taesu, which is target basin of this research, GIS Data was used to divide the basin and establish river network and this allowed to construct input data which can be used to extract basin parameter and estimate TC-K. Also, rain events such as Rusa, Maemi, Ewiniar was selected to conduct runoff estimation. Then, the result from the estimation was underwent TC-K formula to test suitability. Finally, the result of before and after estimation was compared and examined.
As a result of this estimation, TC formulas such as Kirpich, FAA and SCS Log showed fast concentration time and among K formulas, Clark and Linsley showed slow variation. After runoff estimation by using TC-K, the estimated value of peak and total flow were less than actual data and among TC formulas, Izzard formula showed the least TC.
The test of suitability after statistical analysis showed that the relationship between correlation coefficient from Izzard and Kinematic wave formula and storage constant from Clark and Linsley formula is negligible.
Comparing the runoff result before and after the optimization, Kirpich, FAA and SCS Log formula showed no difference. However, observed value (after optimization) from CCP, Rizha and Kinematic wave formula revealed that there is high relationship. Yet, runoff estimation result (before optimization) of Kerby, Kraven(I) and Kraven(II) has excellence compared to the value of after optimization.